The Anatomy of Video Editing: A Dataset and Benchmark Suite for AI-Assisted Video Editing

TL;DR

This paper introduces a comprehensive dataset and benchmark suite for AI-assisted video editing, focusing on tasks beyond visual effects, with extensive annotations to foster research and innovation in the field.

Contribution

It provides the first large-scale annotated dataset and benchmark for AI-assisted video editing tasks like footage organization and assembly, expanding beyond traditional visual effects.

Findings

Over 1.5 million annotated tags related to cinematography

Baseline methods established for various video editing tasks

Detailed analysis provided for each task

Abstract

Machine learning is transforming the video editing industry. Recent advances in computer vision have leveled-up video editing tasks such as intelligent reframing, rotoscoping, color grading, or applying digital makeups. However, most of the solutions have focused on video manipulation and VFX. This work introduces the Anatomy of Video Editing, a dataset, and benchmark, to foster research in AI-assisted video editing. Our benchmark suite focuses on video editing tasks, beyond visual effects, such as automatic footage organization and assisted video assembling. To enable research on these fronts, we annotate more than 1.5M tags, with relevant concepts to cinematography, from 196176 shots sampled from movie scenes. We establish competitive baseline methods and detailed analyses for each of the tasks. We hope our work sparks innovative research towards underexplored areas of AI-assisted video editing.